Semi-Supervised Dimensionality Reduction of Hyperspectral Image Based on Sparse Multi-Manifold Learning

Hong Huang¹, Fulin Luo¹, Zezhong Ma², Hailiang Feng¹

¹Key Laboratory of Optoelectronic Technique and Systems, Ministry of Education, Chongqing University, Chongqing, China.
²Chongqing Institute of Surveying and Planning for Land, Chongqing, China.

DOI: 10.4236/jcc.2015.311006                

Abstract

In this paper, we proposed a new semi-supervised multi-manifold learning method, called semi- supervised sparse multi-manifold embedding (S3MME), for dimensionality reduction of hyperspectral image data. S3MME exploits both the labeled and unlabeled data to adaptively find neighbors of each sample from the same manifold by using an optimization program based on sparse representation, and naturally gives relative importance to the labeled ones through a graph-based methodology. Then it tries to extract discriminative features on each manifold such that the data points in the same manifold become closer. The effectiveness of the proposed multi-manifold learning algorithm is demonstrated and compared through experiments on a real hyperspectral images.

Keywords

Hyperspectral Image Classification; Dimensionality Reduction, Multiple Manifolds Structure, Sparse Representation, Semi-Supervised Learning

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]	Chen, Y., Nasrabadi, N. and Tran, T. (2013) Hyperspectral Image Classification via Kernel Sparse Representation. IEEE Transactions on Geoscience Remote Sensing, 51, 217-231. http://dx.doi.org/10.1109/TGRS.2012.2201730
[2]	Shao, Z. and Zhang, L. (2014) Sparse Dimensionality Reduction of Hyperspectral Image Based on Semi-Supervised Local Fisher Discriminant Analysis. International Journal of Applied Earth Observation and Geoinformation, 31, 122-129. http://dx.doi.org/10.1016/j.jag.2014.03.015
[3]	Qiao, L., Chen, S. and Tan, X. (2010) Sparsity Preserving Projections with Applications to Face Recognition. Pattern Recognition, 43, 331-341. http://dx.doi.org/10.1016/j.patcog.2009.05.005
[4]	Zang, F. and Zhang, J.S. (2011) Discriminative Learning by Sparse Representation for Classification. Neurocomputing, 74, 2176-2183. http://dx.doi.org/10.1016/j.neucom.2011.02.012
[5]	Lu, G.F., Jin, Z. and Zou, J. (2012) Face Recognition Using Discriminant Sparsity Neighborhood Preserving Embedding. Knowledge-Based Systems, 31, 119-127. http://dx.doi.org/10.1016/j.knosys.2012.02.014
[6]	Elhamifar, E. and Vidal, R. (2006) Sparse Manifold Clustering and Embedding. In: Advances in Neural Information Processing Systems, 609-616.
[7]	Lu, J.W., Peng, Y. and Wang, G. (2011) Discriminative Multi-Manifold Analysis for Face Recognition from a Single Training Sample per Person. IEEE International Conference on Computer Vision, 1943-1950. http://dx.doi.org/10.1109/iccv.2011.6126464
[8]	Shi, Q., Zhang, L.P. and Du, B. (2013) Semisupervised Discrimina-tive Locally Enhanced Alignment for Hyperspetral Image Classification. IEEE Transactions on Geoscience Remote Sensing, 51, 4800-4815. http://dx.doi.org/10.1109/TGRS.2012.2230445
[9]	Wang, L.Z., Huang, H. and Feng, H.L. (2012) Multi-Linear Local and Global Preserving Embedding and Its Application in Hyperspectral Remote Sensing Image Classification. Journal of Computer-Aided Design & Computer Graphics, 24, 780-786.
[10]	Song, Y.Q., Nie, F.P. and Zhang, C.S. (2008) Semi-Supervised Sub-Manifold Discriminant Analysis. Pattern Recognition, 29, 1806-1813. http://dx.doi.org/10.1016/j.patrec.2008.05.024
[11]	Cai, D., He, X.F. and Han, J.W. (2007) Semi-Supervised Discriminant Analysis. In: Proceedings of the 11th IEEE International Conference on Computer Vision, 609-616. http://dx.doi.org/10.1109/iccv.2007.4408856